The objective of this proposal is to investigate statistical methods for finding and characterizing genes that interact with environmental factors. In the context of gene finding, we will concentrate on methods that do not require a priori knowledge of a specific model for the trait of interest. In contrast, in the context of gene characterization, we will develop models for the joint dependence of the disease on genes, environment, and GxE interaction in an effort to better understand etiological mechanisms. Specifically, we propose to: Develop and evaluate model-free (nonparametric) linkage analysis methods for finding a gene that interacts with an environmental factor, Compare model-free and model-based (parametric) linkage analysis approaches in the presence of GxE interaction, Develop and evaluate methods of characterizing GxE interestion for a measured gene, and Develop algorithms and associated software for computing sample size and power in the context of gene-finding or gene-characterization studies. Successful completion of these aims will improve our understanding of how GxE interaction influences disease risk, and how we can optimally design studies and analyze data to find and characterize such genes.